YAML default section and per-bench (tau2_bench: 200) both set it; the
runner's generic 8-starved every agent dialog (tau2 reward-0 root
cause). CLI wiring pops the key into args.max_turns before the
generation keys are stripped.
Co-Authored-By: Claude <noreply@anthropic.com>
- config env entries: bfcl_v3->bfcl_mock, tau2_bench->tau2_official;
the eval loop previously POPPED the env key and discarded it -- now
it feeds args.env (CLI --env still works as default/fallback)
- swe_bench_verified images: per-instance names are never published;
the official docker.io layout is repo-level BASE images
(swebench/sweb.eval.x86_64.{repo}) with per-instance images built
on top -- naming corrected (note: the swebench/* namespace is
currently blocked on every reachable CN mirror all the same)
Co-Authored-By: Claude <noreply@anthropic.com>
The adapter dropped reasoning_effort/thinking from the payload, so the
middle rung of the ladder (es reference: full 98.3 / effort_low 94 /
no-think 82.3 on humaneval) was unreachable. Both keys now pass
through; --reasoning-effort {minimal,low,medium,high,max} overrides
the YAML, and config/effort_low.yaml mirrors default.yaml with
reasoning_effort: low for one-command low-thinking runs.
Probe on the endpoint: same question, default = 319 chars reasoning /
262 tok, low = 47 chars / 117 tok -- the server honors it.
Co-Authored-By: Claude <noreply@anthropic.com>
Adding config/sample_counts.yaml made TWO yamls in config/, and the
auto-load rule was 'exactly one yaml' -- so every bench silently lost
its repeats/temperature/max_tokens (humaneval ran once instead of 3).
Manifest renamed to .json, and the rule hardened: a lone yaml still
wins, otherwise default.yaml wins explicitly. Verified: aime 12x +
humaneval 3x repeats active again.
Co-Authored-By: Claude <noreply@anthropic.com>
- sample-counts manifest (config/sample_counts.yaml, harvested from
real runs): uncached benches still show exact numbers in the plan
instead of 'counts when datasets load' -- 'cache+est.' marks the mix
- '--concurrency auto' is now an alias for --auto-concurrency
- Concurrency row shows 'auto (start 8, gate decides)' when the gate
drives, instead of a bare misleading 8
Also verified end-to-end: thinking-mode humaneval rep1/rep2 both
pass 98.8%, matching the es reference runs (98.17/98.78/98.78) on the
same model -- framework alignment holds on the thinking path too.
Co-Authored-By: Claude <noreply@anthropic.com>
syy's es runs: humaneval x3, gpqa x2, aime25/26 x12 (already aligned),
mmlu_pro/longbench_v2 x1 (temp=0 deterministic -- repeats are noise,
and 3x 12k samples is pure cost). temp=1 benches get 3.
Co-Authored-By: Claude <noreply@anthropic.com>
- accurate output structure (summary.xlsx/csv + per-bench xlsx; drop
stale detail.md/summary.md mentions)
- dedicated YAML-config section documenting the keys that actually work
(generation params + repeats + max_input_tokens) and precedence
- perf-stats table: what is always collected vs --perf streaming-only
- FAQ: endpoint probe, context-overflow shrink, thinking-mode notes
(non-stream chat_template_kwargs vs cloud-API param), auto-stream
- config/README.md rewritten to match the flat-key reality (old file
documented a directory scheme + judge/env keys that are not consumed)
Co-Authored-By: Claude <noreply@anthropic.com>
